Sulphuric  Acid,  in  the  Rain-Water  collected  at  Rotkarnsted.  325 
Bearing  this  fact  in  mind,  we  may  probably  take  4-5  lbs. 
per  acre  as  the  best  estimate  we  can  at  present  give  of  the  total 
combined  nitrogen  annually  supplied  in  the  Rothamsted  rain- 
fall. This  is  only  about  two-thirds  as  much  as  the  earlier  results 
indicated  as  due  to  ammonia  and  nitric  acid  alone ; but  it  is 
not  improbable  that,  in  their  case,  a larger  proportion  of  the 
nitrogen  of  the  nitrogenous  organic  matter  was  converted  into 
ammonia,  and  estimated  as  such,  than  in  the  recent  determi- 
nations. 
In  addition  to  the  combined  nitrogen  carried  down  from  the 
atmosphere  in  rain,  we  have  to  consider  any  gain  to  the  soil  or 
to  the  crop  by  direct  absorption  of  ammonia  or  nitric  acid  from 
the  air.  As  far  as  any  gain  from  the  atmosphere  to  the  plant 
itself  is  concerned,  there  is  very  little  direct  experimental 
evidence  on  the  point,  but  such  as  is  available  would  lead  to 
the  conclusion  that  its  amount  is  practically  immaterial.  As  to 
the  amount  of  gain  by  absorption  by  the  soil,  there  is,  unfor- 
tunately, no  direct  or  satisfactory  evidence  at  command.  From 
such  evidence  as  does  exist,  we  are  disposed  to  conclude  that 
with  some  soils  the  amount  will  probably  be  greater,  and  with 
others  less,  than  that  supplied  by  the  rainfall. 
The  Chlorine  in  Rain. 
Determinations  of  chlorine  in  monthly  mixtures  of  rain-water 
have  been  carried  out  at  Rothamsted  since  June  1877 ; the 
results  obtained  up  to  the  close  of  1880  were  published  in  the 
previous  Report  on  Rain  and  Drainage.  The  determinations 
have  been  continued  by  means  of  the  volumetric  method  before 
described,  further  experience  with  this  method  showing  that  it 
gave  results  almost  identical  with  those  obtained  by  the  gravi- 
metric method. 
We  need  not  discuss  in  detail  the  results  obtained  in  in- 
dividual months,  which  will  be  found  in  Table  V.  (p.  326), 
but  turn  at  once  to  the  columns  showing  the  mean  results  of 
the  six  years’  observations. 
In  the  account  given  of  the  earlier  results  of  this  investigation  it 
was  pointed  out  that  the  winter  rainfall  was  far  richer  in  chlorine 
than  the  summer  rainfall ; we  are  now  able  to  take  a step  further, 
and  show  the  general  character  writh  respect  to  chlorine  of  each 
month  in  the  year.  The  minimum  amount  of  chlorine  occurs 
in  the  rain  of  July.  In  August  and  September  there  is  a distinct 
but  not  a very  large  increase  in  quantity.  In  October  and 
November  a great  rise  occurs,  the  quantity  of  chlorine  contained 
in  the  rain  being  three  times  as  large  as  during  the  two  pre- 
ceding months.  After  this  period  of  maximum  there  is  a fall, 
